amdgpu_bo_p..._from_domain(9)

📦 version 7.3.1, in nixpkgs unstable

💅 amdgpu_bo_p..._from_domain(9) — bestie edition

okay so this page lives in the Kernel API Manual, section 9, and its full government name is amdgpu_bo_placement_from_domain (the header just truncates it to amdgpu_bo_p..._from_domain(9) because terminals are tiny and dramatic 🎀). it's a kernel doc, it's giving very driver-dev coded, and we're gonna get through it together. grab a lil snack 🍓

✨ NAME ✨

amdgpu_bo_placement_from_domain — literally just "set buffer's placement", that's the whole vibe 💅

🦋 SYNOPSIS 🦋

void amdgpu_bo_placement_from_domain (struct amdgpu_bo *abo , u32 domain );

it returns void, bestie. nothing comes back. no notes, no receipts, no return value. it just does the thing and leaves 👑

🌸 ARGUMENTS 🌸

  • abo — the amdgpu_bo buffer object whose placement is to be set. this is the main character, the one getting rearranged 🫶
  • domain — the requested domain. like, where do you want this buffer to live, queen? that's the ask 🌈

💖 DESCRIPTION 💖

Sets buffer's placement according to requested domain and the buffer's flags. that's it, that's the function ✨

so real talk: it's not just the domain you pass in — the buffer's own flags get a vote too. two inputs, one placement decision. no cap 🧚

🎀 SEE ALSO 🎀

the kernel file where all this lives is ./drivers/gpu/drm/amd/amdgpu/amdgpu_object.c — go read it, it's right there 👀

and now the friend group, in the exact order they were introduced, no gatekeeping:

like, that's the whole gc right there. a whole squad of amdgpu_bo pages, and every single one said "see also" to the others. iykyk 💀💜

ngl the entire doc is like six lines long and that's iconic of it. short, sweet, does exactly one job. ate and left no crumbs 👑

and that's the tea 🍵 — this page ships out of ./drivers/gpu/drm/amd/amdgpu, dated October 2026, and it's amdgpu_bo_p..._from_domain(9) forever and always. mwah 💋✨🌸

📦 where this page comes from